Audio informed visual speaker tracking with SMC-PHD filter.
Volkan KilicMark BarnardWenwu WangAdrian HiltonJosef KittlerPublished in: ICME (2015)
Keyphrases
- visual information
- audio visual
- visual data
- visual speech
- audio visual speech recognition
- cross modal
- speaker identification
- particle filter
- multimedia
- audio stream
- prosodic features
- real time
- hidden markov models
- low level
- visual features
- multi modal
- compressed video sequences
- unscented kalman filter
- speaker verification
- particle filtering
- visual tracking
- emotion recognition
- speech recognition
- broadcast news
- gaussian mixture model
- mean shift
- speaker recognition
- noise reduction
- object tracking
- visual cues
- acoustic features
- image classification
- video indexing and retrieval
- appearance model
- narrow field of view